Plugins

Adds AMD Radeon Cloud as a model provider in DeepSeek Harness, with reasoning-effort levels declared per model. The provider definition ships in the plugin's own patch layer rather than in your profile.

Elaborates a one-line seed into a concrete, executable prompt behind a semantic fidelity gate (five violation classes, targeted repair, unconditional re-audit); already-precise instructions pass through near-unchanged; bare signals like a number infer their pending anchor from the conversation or refuse to guess; short directives expand inside a degree contract; conversational messages get typo-only touch-up; one-click revert with version history.

Requires the phone app, not just this plugin: the phone half is a separate APK and must be installed first — download it from https://github.com/Zoragyx/dsh-phone-pair/releases — otherwise the computer side has nothing to pair with. This package is the computer half: it enables the upstream remote-web-ui pairing plugin with a Cloudflare quick tunnel and a fixed relay subdomain, so one scan keeps the phone connected from anywhere.

Browser provider for DeepSeek Harness: hand it a web task — open a page, click, type, fill a form, read what is on it — and it drives a real browser step by step, seeing an indexed table of what can be done on the current page instead of raw HTML or screenshots. Each step is one bundled-Jev decision rather than a full model turn (1.1 s median per step versus 2.8–3.6 s for chat-model turns in the repo benchmark), and the main conversation receives only the conclusion, so its context does not grow with the page. Targets come from the page it just read and a changed page is re-read rather than clicked blind, and it runs in the browser that Session owns, leaving your everyday Chrome untouched unless you switch to attach mode.

Four adjustable terse levels plus independent switches for reasoning effort, web search and subagents, on the conversation header and in the Settings page. No output ceiling.

Turns the main model into a dispatcher: six pools (economy/mechanical/main/hard/vision/review) route each subtask to the right model, with ranking and per-route reasoning-effort pins. A context watermark (50k/90k/130k) nudges delegation early and, at the force level, forks a backup session, writes a handover doc, compacts, and continues without losing in-flight subagent work; /ctx-pause, /ctx-resume and /ctx-handover pause, resume or trigger the same machinery by hand. Optional Agent Teams mode adds persistent teammates and a shared task board; /vision lets text-only models handle images via the vision pool; tiered verification sends bigger diffs to a tester and an independent reviewer. Settings UI, READMEs and the conversation / code-comment / document language preferences all cover the same 11 languages (en, zh, zh-TW, ja, ko, de, es, fr, it, pt, ru); the plugin UI follows the DSH app locale unless forced otherwise. Bundles three skills: read-only MySQL/Redis verification, commit-message generation, and a requirements/design document convention.

Registers a real-browser URL-to-text MCP server in a DSH profile: fetch_url and fetch_urls render JavaScript-heavy and anti-bot Chinese pages (Zhihu, CSDN, Juejin, cnblogs) and return clean article text as JSON with per-tier diagnostics, and crawler_status reports whether the browser is available. Needs a one-time Python dependency setup after install.

Run a read-only Dameng DM8 database inspection with 85 checks and produce an HTML report.

Phone client for DeepSeek Harness — no app to install (PWA). Read sessions, background jobs, subagents, goal rounds, workflows and pending approvals, and talk to any session from it: streamed replies, tool-call progress, stop at any time, switch sessions and models. Also ships a GUI sidebar panel. Formerly dsh-taskwatch; the npm package is now dsh-in-hand.

A persistent multi-step task stack for DeepSeek Harness agents: focus_task / focus_complete / read_focus plus a /focus command, with zero automatic prompt injection

Context fullness bands, the per-session compaction threshold derived from the model window, per-step cost computed from cache read and cache write rates, cache-hit, DeepSeek balance with the peak/off-peak tariff, and a one-click handoff. It never calls a model and spends no tokens.

Exposes the models already configured in DeepSeek Harness through an OpenAI-compatible gateway, so other local projects can call them with any standard OpenAI client. Multiple provider candidates share one named group with ordered, balanced or random dispatch; upstream failures before any output switch to the next candidate transparently, and rate-limited requests are retried after the provider's backoff. Per-candidate call statistics (attempts, answers, failures, token/TTFT totals) appear as badges in the settings page, and daily aggregates persist to a local file with a 30-day heatmap and trend view. Until you create a key in the UI, the endpoint accepts calls from anyone who can reach it; keys (stored as SHA-256 hashes) then turn authentication on, and the endpoint stays locked even after the last key is revoked. Exposing the optional LAN listener without a key leaves the port open to the whole network. Runs on DSH 0.1.2-rc.1 through 0.1.7-alpha.1 and later.

Brings SenseNova (日日新) models into dsh with a per-key rotation pool: paste any number of API keys and requests rotate across them, a 429 rate-limit immediately retries on the next key with per-key exponential-backoff cooldown, local RPM pre-throttling avoids the limit instead of reacting to it, and an exhausted pool answers 503 with Retry-After so a 429 never reaches the caller. Ships a built-in model catalogue that works offline and refreshes from /v1/models, plus a settings card with masked keys, per-key probing and removal.

Manage skills and MCP servers from one DSH settings panel. Skills are read from disk: browse, search, preview the body, create, rename and delete, with a backup written before every removal. Global MCP servers are the real cordis.patch.yml loader rows; per-project entries mount only for sessions in that project. Each expanded card lists the tools that server actually registered, and a per-server allow/deny rule keeps only the ones you need — a blocked tool never enters the tool registry, so it costs no context on every turn. Every MCP card carries a connection status tag (connected with a tool count, retrying, gave up, activation failed) plus a one-click reconnect, so a server that fails at startup is visible instead of silently missing its tools. Imports existing configs from 26 other MCP clients.

DeepSeek Harness plugin: tells the model its real, host-measured context occupancy every turn — and keeps compaction summaries from carrying false "context is nearly exhausted" claims.

Billing-period and balance pill under the DSH composer: it shows the account balance and the current conversation cost alongside the peak/off-peak state and a countdown to the next switch. The peak/off-peak rules cover China statutory holidays and weekends, and the conversation cost is measured as the decrease in account balance between the plugin's 60-second idle balance refreshes.

Create Blender product scenes in DSH with editable recipes, model imports, previews and PNG/MP4 delivery; includes workbench and agent configuration.

Cross-session long-term memory backed by a local Go service (~50MB RAM, SQLite-vec + verify-augmented claims): zero-lag recall injection on every user message, session-end transcript ingest, and memory_search / memory_logs / memory_ingest tools. Install with `dsh plugin --profile web add @djasdh/interest-memory-dsh-bridge`. Protocol bridge only: the Go service is not installed by `dsh plugin add` — install and start it per the repository README first; until it runs, the tools silently return nothing.

Adds hover Recall / Delete / Delete-onward actions to every message row: removes content from the model context and the transcript together via durable append-only tombstones, with an empty-content guard so old deletions can never break strict providers. Tested on the DSH 2.0 desktop.

Bidi-safe RTL rendering for mixed Arabic and English text, a full Arabic interface pack (3,307 strings) registered through the DSH locale service, and three thmanyah cuts — Sans for the interface, Serif Text for reading, Serif Display for headings — placed behind whatever font the app itself is set to.

SenseNova Token Plan credit panel rendered as a config card on the Harness Plugins page: per-pool quota windows (5h and weekly), grant balance and per-model consumption from the SenseNova console API, with in-panel login and automatic token renewal. Three opt-in switches, off by default, register SenseNova models and the Xiaohuanxiong upstream as LLM providers and expose an image-generation tool to the agent.

Real-time disaster alerts for DeepSeek Harness: Japan earthquakes (P2PQuake) and JMA EEW, tsunami and weather telegrams, global earthquakes (EMSC/USGS) and NOAA tsunamis, mainland China earthquakes (CENC via Wolfx) plus nmc.cn rainstorm and geological-hazard warnings, and US/Canadian NWS and ECCC flood, rainfall and storm-surge warnings — matched against your watch regions and thresholds.

DSV4.1 reasoning support with multimodal context and bounded artifact acceptance for DeepSeek Harness

Local TXT reader in a right-sidebar tab plus a spoiler-safe AI reading companion: the model only sees what you have already read — the current chapter, the tail of the previous chapter, and a per-book background file it grows as you go. Excerpts, thoughts and the companion replies become structured Markdown notes exportable to Obsidian; raw text unlocks only for a book you mark as finished.